An approach to optimization of airport taxiway scheduling and traversal under uncertainty
Airport runways and taxiways have been identified as a key source of system-wide congestion and delay in the over-strained commercial air traffic system. To combat this growing problem, we present a novel approach for taxiway scheduling and traversal. Aircraft must traverse a taxiway, represented by a graph, from gates to their respective runways and arrive at their scheduled times while adhering to safety separation constraints. We describe a combinatorial mixed-integer linear program to determine the push-back time windows, aircraft speeds, stopping times, and in particular, traversal paths for a given graph and an imposed flight schedule as part of a single optimization problem. Safety and scheduling constraints are made robust to probabilistic deviations from the prescribed schedule and aircraft motion, and multiple objective functions are considered to examine the trade-off between taxi times and the probability of safety separation violation. Several scenarios are presented to demonstrate improvements gained from the method and possible uses for this approach.